gave a talk last weekend on how voice AI works. the stt→llm→tts pipeline, why you get ~800ms before a conversation feels broken, and how orchestrators like pipecat/livekit run the whole loop
live demo ran on @pipecat_ai with @smallest_AI's Pulse + Lightning models (the stack i
149 new voices are now live on Lightning V3.1 Pro.
Expanding your voice library across even more languages:
• 67 voices added across 9 Indian languages
• 21 voices across Asian & Middle Eastern languages
• 61 voices across 11 European languages
Available now in the
Smallest Pulse just tied #1 in the most rigorous streaming STT benchmark we've seen.
Here's what the data says.
An independent benchmark put 6 streaming STT models through 5 production conditions - accented speech, telephone audio, background noise, medical vocabulary, and
A lead fills your form → they get a voice note greeting them by name on WhatsApp.
A deal closes in the CRM → the win is announced out loud in the sales channel.
A blog post goes live → the audio version publishes itself.
Get all of this done with our n8n workflow: POST your